hospital-readmission

ClinicalBERT: Modeling Clinical Notes and Predicting Hospital Readmission (CHIL 2020 Workshop)

Hospital readmissions prediction models

Prediction of a readmission for a patient based on the Electronic Health Records (EHR) data. This project was done as part of a timed challenge with a time limit of 3 hours to work on this dataset. So, it is just a preliminary model using XGBoost algorithm with some basic data exploration for data processing.

End-to-end healthcare analytics project analyzing 30-day hospital readmission rates across 2,245 U.S. hospitals using CMS HRRP FY2026 data | SQL Server · Python · Power BI | $402M excess cost identified

Predicting 30-day hospital readmissions for diabetic patients using ML — Logistic Regression, Random Forest & XGBoost on UCI Dataset

Machine learning project to predict 30-day hospital readmission risk using healthcare data, class imbalance handling, threshold tuning, and patient-safety-oriented evaluation.

End-to-end ML pipeline predicting 30-day hospital readmission for diabetic patients. XGBoost + SMOTE on 101,766 UCI records. ROC-AUC 0.748 | Recall 0.682. Final Project — Data Science.

Machine learning project to predict 30-day hospital readmission risk for diabetic patients using XGBoost, CatBoost, LightGBM, threshold tuning, and Streamlit.

Hospital readmission prediction ML pipeline in R — 6 algorithms (LASSO, Random Forest, SVM, KNN, Naive Bayes, Decision Tree) on 69,984 diabetic patients using Tidymodels + SMOTE

Machine learning models to predict hospital readmission risk using clinical data

🏥 Predictive analytics command center for 30-day hospital readmission risk using ML, Streamlit, Power BI and SQL

Predizione della riammissione ospedaliera a 30 giorni in pazienti diabetici: pipeline ML end-to-end e un esperimento sul peso del protocollo di validazione rispetto alla scelta del modello.

AI-powered healthcare web application for predicting patient readmission risk and insurance claim amounts using machine learning.

Leakage-safe prediction of 30-day hospital readmission among patients with diabetes using calibrated, interpretable machine-learning models in Python.

End-to-end deep learning (DNN/CNN/embeddings) for 30-day hospital readmission on the UCI Diabetes 130-US-Hospitals benchmark, under an honest patient-grouped evaluation protocol. Companion to leakage-lens.

Interpretation-focused analysis of 30-day ICU readmission using the MIMIC-III clinical database, with emphasis on cohort definition, data preprocessing, and clinical context.

XGBoost + LightGBM ensemble predicting 30-day hospital readmission risk on 101K patient records. AUC-ROC 0.71 with SHAP explainability and fairness audit.

Machine learning pipeline predicting 30-day hospital readmissions using XGBoost, Random Forest, and Logistic Regression on the UCI Diabetes 130-US Hospitals dataset

A production-grade clinical analytics pipeline examining 30-day hospital readmission risk Built with PostgreSQL · Supabase · Tableau Public · Advanced SQL